Answer to Question #9635 Submitted to "Ask the Experts"
Category: Radiation Basics
The following question was answered by an expert in the appropriate field:
Why would I have a gamma exposure rate on the order of 0.139 nGy s-1 at 30 cm from a 45-year-old Po-Be neutron source (original activity 2.59 x 1011 Bq)? Is there an impurity in the alloy that would have been activated in the test reactor that it was used in? The reactor has been shut down since 2002.
The usual method of production of 210Po, in the multicurie quantities often used in portable neutron sources, involves irradiating stable 209Bi in a high thermal neutron fluence, typically available in a nuclear reactor, to yield 210Bi, which decays with a five-day half-life to 210Po. The 210Po is then separated from the bismuth and mixed with beryllium to make the (a,n) source. The 210Po does emit a very low-yield gamma ray in its decay, but the 138-day half-life precludes any contribution of this gamma ray to measured exposure rate after the 45-year interval since the source was fabricated. The bismuth that is irradiated is generally quite pure, but depending on the purity of the bismuth used, there could possibly have been some contaminants present that might have yielded some additional activation products. It is unlikely, however, that these would carry through in the 210Po separation, be incorporated into the fabricated source, and persist for the length of time of concern here.
I assume that the source you refer to was used as a start-up source for the test reactor and was doubly encapsulated in stainless steel. It is true that some constituents in the stainless steel would get activated when the source was in proximity to the reactor core as neutrons were being generated. The likely metals used in making most stainless steel could produce some gamma-emitting, neutron activation products with long enough half-lives to make them measurable nine years after removal from the reactor. For example, stable 54Fe could undergo a fast neutron (n,p) reaction to yield 312-day half-life 54Mn, which emits 835 keV gamma rays. Similarly, stable 60Ni could undergo an (n,p) reaction to generate 60Co, a well-known gamma emitter. One fact that mitigates against these products is that the fast neutron fluence rates at the source location may have been relatively low, although the source storage and use history and the reactor characteristics are not clear.
Another possible source of the detected gamma radiation could be 60Co produced by thermal neutron activation of stable cobalt present in the stainless steel; cobalt-containing stainless steels are relatively uncommon, and I don’t think that such would have been used for encapsulating the 210Po, but it is possible. We can look at the implications of that possibility.
To obtain a reading of 0.139 nGy s-1 at 30 cm would imply an approximate 60Co-equivalent activity of 1.3 x 105 Bq. You note that the reactor ceased operation in 2002, nine years ago. The polonium source would presumably have become too weak to use after several 210Po half-lives, implying that within a few years after obtaining the source it may not have been used again as a start-up source. You do not specify whether, after that time, the polonium source continued to be stored in the reactor over the entire lifetime of the reactor. For purposes of this estimation, I shall assume that to have been the case. If the activity actually were 60Co, the 1.3 x 105 Bq would then translate to about 4.3 x 105 Bq at a time nine years earlier when the reactor shut down for good. If the source had been used originally as a start-up source and then stored in the reactor, it would have spent most of its time somewhat removed from the reactor core but may still have experienced a moderate and extended fluence rate of leakage neutrons. I won’t provide the details here, but if we assume that the stainless steel contained cobalt at the maximum likely percentage of 0.2 percent (as sometimes found in type 348H stainless), and if we assume that the encapsulation weighed 150 grams (I’m just guessing here since I don’t have specific source details), we can show that an average steady thermal neutron fluence rate of between 106 and 107 neutrons cm-2s-1 would have been sufficient to produce the 4.3 x 105 Bq of 60Co.
I do not know anything about the specific test reactor to which you refer, but in-core thermal neutron fluence rates of 1013 to 1014 cm-2 s-1 would not have been unusual for a number of such reactors, and leakage fluence rates at modest distances from the core could well have been in the range of 106 to 107. Naturally, this is all a simplified and presumptive approach; it is intended only to ascertain whether the measured dose rate and back-calculated 60Co-equivalent activity were sensible or feasible.
I don’t think it would be productive for me to make further conjectures as to the nature of the source of the presumed gamma radiation that you have measured. If you have a sufficient interest to pursue this further, the best approach to identification of the responsible radionuclide(s) would be to perform gamma spectrometry to determine the photon energies involved. From that identification it would be an easier matter to specify the likely source of the radionuclide(s).
Sorry I cannot be more definitive in my response.
George Chabot, PhD, CHP